Printer Friendly

Cutaneous metastasis: a meta-analysis of data. (Original Article).

Background: Previous studies of the incidence and the most common causes of cutaneous metastasis have not led to a consensus. We compiled data from many retrospective studies and from patient data registries and autopsies to increase the total number of cases available for a larger analysis of this subject. This study was conducted to gain a better understanding of the true incidence of cutaneous metastasis, the tumors most commonly involved in this presentation, and the locations of such lesions.

Methods: A meta-analysis of cutaneous metastases from patient tumor registries and autopsic studies was performed.

Results: The overall incidence of cutaneous metastasis is 5.3%. The most common tumor to metastasize to the skin is breast cancer. The chest is the most common site of cutaneous metastasis. Every practitioner should be highly suspicious of acute-onset, persistent, firm papulonodules, especially when they develop on the chest.

Conclusion: This meta-analysis greatly increases the total number of cases available for the analysis of cutaneous metastases and provides a better overall view of this topic than was previously possible.


Although the general consensus in the medical literature is that cutaneous metastases from visceral malignancies are uncommon, all studies conducted to date have reached different conclusions on their incidence. (1-11) Estimated incidence rates have ranged from 0.7% in a series of 865 autopsies reported by McWhorter and Cloud (1) to 9% in an autopsic study of 7,518 cases described by Spencer and Helm. (11) Rosen, (12) in a previous review of five large studies, showed the overall incidence to be approximately 2%. We have examined nine studies of cutaneous metastases, thereby vastly increasing both the number of cases and the geographic areas represented by the institutions involved. This meta-analysis was performed to obtain a better understanding of the true overall incidence of cutaneous metastasis, which cancers most commonly metastasize to the skin, and which skin sites are prone to demonstrate this phenomenon. Such information is critical to the busy primary care physician.

Patients and Methods

We gathered data from nine large-scale studies of cutaneous metastases. For the determination of the overall incidence of cutaneous metastasis, 7 studies comprising a total of 22,297 patients were used. After removing cases of malignant melanoma, leukemia, and lymphoma wherever possible, the total number of cases remaining was 20,380. Each of these seven studies examined a large number of cancer patients; six of them used autopsic data, and one used retrospective chart review. These studies calculated the incidence of cutaneous metastasis on the basis of the total number of cancer patients observed. Only four of these seven studies further analyzed the incidence of cutaneous metastasis according to primary tumor type. These four studies were then used to determine the overall incidence of cutaneous metastasis from each tumor type. Three of the nine large-scale studies--one of which was in the previous group of seven--7were used to determine the sites of metastasis. For this part of the analysis, the total num ber of metastases was 1,013. Data gathered from patients with malignant melanomas were excluded wherever possible. The percentages of metastases to each site were calculated as a percentage of the total number of metastases recorded, not of the total number of cases, because each patient may have had more than one metastasis and because more than one site may have been involved.


Data obtained from these studies were compiled and analyzed. Cutaneous metastases were found in 1,080 of the total pool of 20,380 cancer patients. The overall incidence of cutaneous metastasis from visceral neoplasia was thus determined to be 5.3%. This value deliberately excluded cases of malignant melanoma, leukemia, and lymphoma. The tumor with the highest incidence of cutaneous metastasis was breast cancer, which was present in 457 of 1,903 cases, an incidence of 24.0%. Lung cancer, colorectal cancer, renal cancer, ovarian cancer, and bladder cancer all had similar rates of between 3.4 and 4.0% for cutaneous metastasis. Prostate cancer had a 0.7% incidence of metastasis to the skin (Table I). The most common site for metastasis was the chest, which was involved in 28.4% of the total number of metastases. The second most common site was the abdomen, at 20.2%. In descending order of incidence, other involved sites were the extremities (12%), the neck (11%), the back (11%), the scalp (7%), the pelvis (6%), a nd the face (5%).


Cutaneous metastasis is a relatively uncommon manifestation of visceral malignancy. It most often occurs late in the course of disease but also may be the presenting sign of underlying cancer. Cutaneous metastasis arising from visceral malignancies has no specific appearance. These lesions are often described as either cutaneous or subcutaneous nodules and as flesh-colored to pink or violaceous, and often the patient is asymptomatic (Figs. 1 and 2). However, many reports of non-nodular metastases exist. Renal cell carcinoma has presented as a cutaneous horn. (13) Pyogenic granulomalike lesions have been described in connection with breast cancer and renal cell carcinoma. (14) An adenocarcinoma of unknown origin was manifested as facial lymphedema caused by lymphatic obstruction. (15) Carcinoma erysipelatoides is a cutaneous metastatic process characterized by erythema, edema, warmth, and tenderness secondary to lymphatic vessel occlusion and has been associated with breast cancer, adenocarcinoma of unknown or igin, and other tumors. (9,16,17) Alopecia neoplastica, a rare presentation of cutaneous metastasis often associated with breast cancer, manifests itself clinically as areas of scarring alopecia. (18-20)

There is no agreement in the literature regarding the exact incidence of cutaneous metastasis. Variability in the estimates of the exact incidence as determined by previous authors can be explained by the intrinsic error of the studies, according to Rosen. (12) Autopsic studies are limited by the potential to miss micrometastasis and may not account for metastases that might have responded to therapy before death. Retrospective studies may miss cases in which cutaneous metastasis of a visceral tumor was evident and therefore not biopsied.

We found the overall incidence of cutaneous metastasis to be 5.3% using information from a total of 20,380 cases. As previously described, there are limitations to the methods of collecting data for these studies, and indeed we recognize the limitations of our own analysis. First, one of the studies (6) is a retrospective chart review rather than an autopsic series. Thus, all of the data used were not collected in an identical manner. However, we contend that this study still provides the same information as the other studies and is therefore valid. Second, each study has its own intrinsic yet unmeasurable error. Our compilation of these studies may have summated that error as well. On the other hand, the sheer aggregate volume of cases may equally have diluted out various sources of error.

The tumor most commonly found to metastasize to skin is generally considered to be breast cancer. In Reingold's study, (7) which contained no women, lung cancer accounted for the majority of the cases. In our analysis, breast cancer was found to be the tumor that most frequently metastasized to the skin, accounting for 24% of cases. Unfortunately, not all previously published studies differentiated true distant metastasis from local chest wall involvement by direct extension of tumor, so the true incidence may be lower than that calculated on the basis of the available data. Lung cancer, colorectal cancer, renal cancer, ovarian cancer, and bladder cancer all metastasized at a rate of approximately 4%. Prostate cancer, which has an extremely high prevalence, was found to metastasize in only 0.7% of cases. All other types of cancer in this collection of studies were found to metastasize at even lower rates when the data were pooled.

The most common site of metastasis was found to be the chest, which was involved in 28.4% of metastases. One possible reason for the high number of metastases to the chest is the high body surface area of that region. Another potential reason is that most cutaneous metastases tend to occur near the primary neoplasm. (12) Given the high numbers of breast and lung cancers, the propensity for metastasis to this location seems reasonable. There has been considerable discussion in the literature about the relatively high incidence of metastases to the scalp, especially in patients with renal cell carcinoma. (8,10,21) In a recent review of 10 cases of cutaneous metastasis in renal cell carcinoma, the scalp was found to be involved in 50%. (21) We found scalp involvement in 6.9% of all metastases, regardless of histologic type. Speculation regarding why the scalp seems to be involved so frequently includes the high degree of vascularity, immobility, and the warmth of the region. (2,22)


We have found the overall incidence of cutaneous metastasis to be 5.3% by compiling data reported in numerous previous studies, comprising both autopsic series and tumor patient registries. The most common tumor to metastasize to the skin was breast cancer, with an incidence of 24%. The most common site of involvement was the chest, which was involved in 28.4% of cutaneous metastases. Although this meta-analysis is not without flaws, we think that it increases the understanding of the statistical probabilities regarding the various parameters related to cutaneous metastasis from visceral malignancy.
Table 1

Incidence of tumors most commonly found to metastasize to skin

 No. of cutaneous metastases/total
 no. of autopsic cases

Study Breast Renal

Abrams, 1950 (4) 31/167 0/13
Enticknap, 1952 (3) 4/45 0/34
Spencer and Helm, 1987 (11) 185/699 6/225
Lookingbill et al, 237/992 12/183
 1990, (6) 1993 (5)
Total no. of patients (%) 457/1,903 (24.0%) 18/455 (4.0%)

 No. of cutaneous
 metastases/total no. of
 autopsic cases

Study Ovarian Bladder

Abrams, 1950 (4) 1/16 1/28
Enticknap, 1952 (3) 1/64 0/2
Spencer and Helm, 1987 (11) 10/337 7/314
Lookingbill et al, 14/272 13/232
 1990, (6) 1993 (5)
Total no. of patients (%) 26/689 (3.8%) 21/576 (3.6%)

 No. of cutaneous metastases/total
 no. of autopsic cases

Study Lung Colorcetal

Abrams, 1950 (4) 7/196 1/164
Enticknap, 1952 (3) 1/160 8/205
Spencer and Helm, 1987 (11) 21/1,223 18/772
Lookingbill et al, 60/1,018 27/450
 1990, (6) 1993 (5)
Total no. of patients (%) 89/2,597 (3.4%) 54/1,591 (3.4%)

 No. of
 al no. of
 autopsic cases

Study Prostate

Abrams, 1950 (4) 0/24
Enticknap, 1952 (3) 0/2
Spencer and Helm, 1987 (11) 0/426
Lookingbill et al, 5/310
 1990, (6) 1993 (5)
Total no. of patients (%) 5/762 (0.7%)

Accepted June 7, 2002.


(1.) McWhorter JE, Cloud AW. Malignant tumors and their metastases: A summary of the necropsies of eight hundred sixty-five cases performed at the Bellevue Hospital of New York. Ann Surg 1930;92:434-443.

(2.) Gates O. Cutaneous metastases of malignant disease. Am J Cancer 1937; 30:718 -730.

(3.) Enticknap JB. An analysis of 1,000 eases of cancer with special reference to metastasis. Guy's Hasp Rep 1952;101:273-279.

(4.) Abrams H, Spiro R, Goldstein N. Metastases in carcinoma: Analysis of 1,000 autopsied cases. Cancer 1950;3:74-85.

(5.) Lookingbill DP, Spangler N, Helm KF. Cutaneous metastases in patients with metastatic carcinoma: A retrospective study of 4020 patients. J Am Acad Dermatol 1993;29:228-236.

(6.) Lookingbill DP, Spangler N, Sexton FM. Skin involvement as the presenting sign of internal carcinoma: A retrospective study of 7316 cancer patients. J Am Acad Dermatol 1990;22:19-26.

(7.) Reingold IM. Cutaneous metastases from internal carcinoma. Cancer 1966;19:162-168.

(8.) Connor DH, Taylor HB, Helwig EB. Cutaneous metastasis of renal cell carcinoma. Arch Pathol 1963;76:339-346.

(9.) Brownstein MH, Helwig EB. Spread of tumors to the skin. Arch Dermatol 1973;107:80-86.

(10.) Brownstein MH, Helwig EB. Metastatic tumors of the skin, Cancer 1972;29: 1298-1307.

(11.) Spencer PS, Helm TN. Skin metastases in cancer patients. Cutis 1987;39:119-121.

(12.) Rosen T. Cutaneous metastases. Med Clin North Am 1980;64:855-900.

(13.) Peterson JL, McMarlin SL. Metastatic renal-cell carcinoma presenting as a cutaneous horn. J Dermatol Surg Oncol 1983;9:815-818.

(14.) Hager CM, Cohen PR. Cutaneous lesions of metastatic visceral malignancy mimicking pyogenic granuloma. Cancer Invest 1999;17:385-390.

(15.) Jang KA, Choi JH, Sung KJ, Moon KC, Koh JK. Cutaneous metastasis presenting as facial lymphedema. J Am Acad Dermatol 1998;39:637-638.

(16.) Alonso-Llamazarcs J, De Pablo P, Ballestin C, Guerra A, Iglesias L. Cutancous metastasis from a presumed signet-ring cell carcinoma in a 10-year-old child. Br J Dermatol 1998;138:145-149.

(17.) Hazelrigg DE, Rudolph AH. Inflammatory metastatic carcinoma: Carcinoma erysipelatoides. Arch Dermatol 1977;113:69-70.

(18.) Baum EM, Omura EF, Payne RR, Little WP. Alopecia neoplastica: A rare form of cutaneous metastasis. J Am Acad Dermatol 1981;4:688-694.

(19.) Martin J, Ross JB. Alopecia totalis as a presentation of cutaneous metastasis (alopecia neoplastica). Int J Dermatol 1983;22:487-489.

(20.) Mallon E, Dawber RP. Alopecia neoplastica without alopecia: A unique presentation of breast carcinoma scalp metastasis. J Am Aced Dermatol 1994;31:319-321.

(21.) Dorairajan LN, Hemal AK, Aron M, Rajeev TP, Nair M, Seth A, et al. Cutaneous metastases in renal cell carcinoma. Urol Int 1999;63:164-167.

(22.) Fay T, Henry GC. Correlation of body segmental temperature and its relation to the location of carcinomatous metastasis: Clinical observations and response to methods of refrigeration. Surg Gynecol Obstet 1938;66:512-514.


* Internal malignancy may spread to the skin and may even manifest initially as cutaneous metastasis.

* Approximately 5% of visceral malignancies eventually become cutaneous metastases.

* Breast cancer is the most common malignancy to involve the skin, whereas prostate cancer is the least likely malignancy to do so.

* The chest and abdomen are the most common cutaneous sites of cutaneous metastases, whereas the face and scalp are the least common.

From the Department of Dermatology, Baylor College of Medicine, Houston, TX.

Reprint requests to Ted Rosen, MD, Department of Dermatology, Baylor College of Medicine, 6560 Fannin Avenue, Suite 802, Houston, TX 77030. Email:

Copyright [c] 2003 by The Southern Medical Association 0038-4348/03/9602-0164
COPYRIGHT 2003 Southern Medical Association
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2003, Gale Group. All rights reserved. Gale Group is a Thomson Corporation Company.

Article Details
Printer friendly Cite/link Email Feedback
Author:Rosen, Ted
Publication:Southern Medical Journal
Geographic Code:1USA
Date:Feb 1, 2003
Previous Article:Her middle name was crystal.
Next Article:McCarthy M.

Related Articles
Understanding Cancer's Spread.
Hepatocellular cancer metastatic to the zygoma: Primary resection and immediate reconstruction. (Original Article).
Spinal metastasis as the initial manifestation of a nonsecretory glucagonoma. (Case Report).
Cystosarcoma phyllodes metastatic to the mandible: report of a rare case and literature review. (Original Article).
A case of renal cell carcinoma metastatic to the nose and tongue. (Original Article).
Dermal metastases from a visceral primary.
Skin metastasis in a previously irradiated field from squamous cell carcinoma of the cervix.
Cutaneous metastases of signet cell carcinoma of the rectum without accompanying visceral involvement.
Invasive metastatic skin cancer in the background of chronic lymphocytic leukemia.
The role of endoscopic sinus surgery in the diagnosis and treatment of metastatic orbital carcinoid tumors.

Terms of use | Copyright © 2017 Farlex, Inc. | Feedback | For webmasters